← Back to community benchmarks
Qwen3.6-27B-oQ6-mtp
Performance
4k
tokens
433.8
PP tok/s
18.7
TG tok/s
9442
TTFT (ms)
24.4
Peak mem (GB)
Hardware
Chip
M5 Pro (20c)
Memory
64 GB
GPU Cores
20
Software
oMLX
v0.3.9.dev2
macOS
macOS 26.5
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 396.8 | 19.4 | 23.0 GB | view |
| 4k | 433.8 | 18.7 | 24.4 GB | current |
| 8k | 372.3 | 16.3 | 25.4 GB | view |
| 16k | 333.3 | 15.5 | 26.9 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 19.4 | 1.00× |
| 2× | 20.4 | 1.05× |
| 4× | 22.9 | 1.18× |
| 8× | 23.6 | 1.22× |